Goals of this continuation application are to identify and characterize two subtypes of developmental phonological (speech) disorders currently of unknown origin. Research to date suggests that the most prevalent form may be genetically transmitted; it is not clear whether there is an autonomous phonological form or whether the genotype always also codes for some level of language disorder. A second subtype may be a phonological disorder due to concomitants of early recurrent otitis media with effusion. The research questions to be addressed in this continuation application are (a) what are the phenotype markers or distinctive speech or prosody-voice patterns for each of the putative subtypes; (b) how do gender, language status, and intelligibility interact with speech and prosody-voice status; and (c) are the relevant phonological data available in the conversational speech samples of probands and affected relatives at different periods of the life span. The internal and external validity of findings from the five-year research plan will be based on data (several numbers are approximate) from 1,200 speech samples obtained from 950 subjects ranging in age from 3 to 60 years. Eight cross-sectional and longitudinal data sets will include (a) a database of 312 children with normal and disordered phonological development, (b) one prospective and three follow-up study series of 130 children with developmental phonological disorders, and (c) three collaborative studies providing cross-sectional and longitudinal information on 516 children and adults with and without histories of a developmental phonological disorder. Data reduction will use measures and conversational speech sampling procedures designed for multivariate statistical analyses of phonetic, phonologic, prosodic, and language variables. The multivariate analyses will test 12 hypotheses divided into two topic areas. Phenotype Studies will test (a) eight hypotheses about concurrent and longitudinal associations among gender, language status, and speech and prosody-voice status in persons with and without histories of a developmental phonological disorder and (b) two hypotheses about the strength of concurrent and cross-lagged correlations among speech and prosody-voice in conversational speech, measures of speech awareness, and challenging production tasks. Otitis Media Studies will test two hypotheses about the speech and prosody-voice profiles of children with positive and negative histories of early recurrent otitis media with effusion. The development and validation of classification systems was identified as a specific research target in the most recent NIH national task force report on research needs in communicative disorders (NIH, 1989). Findings from this proposal should have significant impact on theoretical issues and clinical needs in developmental phonological disorders. With a prevalence of 7.5% in 3- to 11-year-old children, phonological disorders comprise the second largest group (after learning disabilities) among the approximately 4.8 million school children for whom special educational resources must be allocated (U.S. Department of Education, 1992). Results should contribute to current diagnostic, prognostic, and intervention procedures, and ultimately, to methods for primary, secondary, and tertiary forms of prevention (ASHA, 1991).